Problems I see with that... You'd need to cover from ELF up to FIR.

I would make the statement that NIR up through gamma/xrays is not being generated by the human body.

With that, we're looking at around 10THz of bandwidth for a full spectrum. At 8bit IQ acquisition, we're looking at 20TB/s... And if that's assuming you have ideal transmitters and receivers for the whole set of bands.

You want to also transmit to simulate if you find an area that seems to have a response. The test subject should feel something if that's the case.